Coenzyme A-mediated transacylation of sn-2 fatty acids from phosphatidylcholine in rat lung microsomes

458

Biochimica et Biophysics Acta 875 (1986) 458-464 Elsevier

BBA 52126

Coenzyme A-mediated transacylation of ~2-2 fatty acids from phosphatidylcholine

in rat lung microsomes

J.G. Nijssen and H. van den Bosch Lahorato~v OJ Bmchemistrv,

State Unioersrty of Utrechr, Padualuan b: NL-3584 (Received (Revised manuscript

Key words:

CoA-mediated

CH Utrecht (The Netherlands)

January 31st, 1985) received October 25th. 1985)

acyl transfer;

Rat lung microsome;

Phosphatidylcholine

Evidence was obtained for a CoA-dependent transfer of linoleate from rat lung microsomal pbospbatidylcholine to lysophosphatidylethanolamine without the intervention of a Ca” -requiring phospholipase A 2 activity and ATP. To study this CoA-mediated transacylation process, microsomes were prepared in which the endogenous phosphatidylcholine was labeled by protein-catalyzed exchange with phosphatidylcholines containing labeled fatty acids in the sn-2position. The apparent K, for CoA in the transfer of arachidonate from phosphatidylcholine to I-acyllysophosphatidylethanolamine was 1.5 FM. At saturating lysophosphatidylethanolamine concentrations, the transacylation was linear with the amount of microsomal protein, i.e., a fixed percentage of the labeled fatty acid was transferred independent of the amount of microsomal protein. A maximal transfer of 12.2% for arachidonate and 2.0% for linoleate from the respective phosphatidylcholines to lysophosphatidylethanolamine was observed in 30 min. With I-acyl-2-l l-‘4C]arachidonoylphosphatidylcholine as acyl donor, lysophosphatidylethanolamine was the best acceptor followed by lysophosphatidylglycerol and lysophosphatidylserine. Lysophosphatidate barely functioned as acceptor. These data provide further evidence for the widespread occurrence of CoA-mediated transacylation reactions. The arachidonate transacylation from phosphatidylcholine to other phospholipids in lung tissue may contribute to the low level of arachidonate in pulmonary phosphatidylcholine.

Introduction It has been well-established that arachidonic acid is stored in cellular phosphoglycerides esterified exclusively at the sn-2-position [l]. Rather than by de novo synthesis, the assembly of the arachidonoyl species of phospholipids is thought to proceed via a deacylation-reacylation pathway [2,3]. The last step in the reaction sequence involves the acylation of I-acyllysophosphoglyceride and is catalyzed by acyl-CoA : l-acyllysophosphoglyceride acyltransferase. The acyl-CoA : lysophosphoglyceride acyltransferase-mediated reaction is considered to be reversible, as first demonstrated by Dawson and Irvine [4]. Consequently, this reac0005-2760/86/$03.50

0 1986 Elsevier Science Publishers

tion can lead to a direct, ATP-independent, formation of acyl-CoA ester. Since the enzyme appears to exhibit a high preference for arachidonate (reviewed in Refs. 1 and 5) the reverse reaction could thus provide a means for a specific cleavage of arachidonate from glycerophospholipids. In murine thymocytes, a CoA-dependent cleavage of arachidonic acid from phosphatidylcholine could be monitored. The sole lysophospholipid that was able to act as an acceptor for arachidonate was I-acyl-2-lysophosphatidylethanolamine, thus leading to a transacylation of arachidonate exclusively from phosphatidylcholine to phosphatidylethanolamine [6,7]. A bidirectional CoA-mediated transacylation of arachidonate between phos-

B.V. (Biomedical

Division)

459

phatidylcholine and phosphatidylethanolamine as well as a unidirectional transfer from phosphatidylinositol to either I-acyl-2-lysophosphatidylethanolamine or 1 -acyl-2-lysophosphatidylcholine was demonstrated by Flesh et al. [8] in macrophages derived from murine bone marrow. Furthermore, it was established by these authors that oleic acid and linoleic acid could be transferred as well, though to a lower extent than arachidonic acid. Two distinct transacylase activities appear to be present in human platelets, both catalyzing the synthesis of arachidonate-containing phospholipids from the corresponding lysophospholipids. Without CoA, arachidonate from phosphatidylcholine was shown to be incorporated only into 1 -acylor I-alkenyl-2-lysophosphatidylethanolamine with a high preference for the plasmalogen form [9]. However, in the presence of a CoA, not only lysophosphatidylethanolamine but also lysophosphatidylserine and lysophosphatidylinositol became acylated [lo]. The latter authors could also demonstrate that the transacylase activity had a high preference for arachidonate and for phosphatidylcholine as acyl donor. Remodeling of pulmonary phosphatidylcholine is considered to proceed via a classical deacylation-reacylation mechanism [12-211 involving phospholipase A >-catalyzed removal of unsaturated fatty acids from the sn-2-position followed by acylation of the resulting lysophosphatidylcholine with acyl-CoA. In the previous paper [ll], we obtained evidence for the CoA-dependent transfer of linoleate from phosphatidylcholine to lysophosphatidylethanolamine without intervention of Ca’+-requiring phospholipase AZ activity and ATP. In this paper, some properties of this CoA-dependent transacylation process are described using microsomes in which the endogenous phosphatidylcholine has been labeled by proteincatalyzed phosphatidylcholine exchange. Materials

and Methods

Materials I-Acyl-2-[l-‘4C]arachidonoylphosphatidylcholine (58 mCi/mmol) and [l-14C]linoleic acid (57 mCi/mmol) were purchased from Amersham, U.K. l-Acyl-2-[l-‘4C]linoleoylphosphatidylcholine was synthesized biochemically from l-acyl-

lysophosphatidylcholine and linoleoyl-CoA as described by Van den Bosch et al. [22] for the synthesis of 1-acyl-2-[ l-l4 Cllinoleoylphosphatidylethanolamine. Phosphatidylcholine-exchange protein as purified from bovine liver [23] was a generous gift of Dr. D. van Loon from this laboratory. Lysophosphatidylserine, lysophosphatidylglycerol and lysophosphatidylethanolamine were prepared by degradation of the corresponding diacylphosphoglycerides with pig pancreas phospholipase AZ (kindly donated by Dr. H.M. Verhey) and purified on silica gel type H thin-layer plates. Egg phosphatidylcholine and phosphatidylglycerol prepared therefrom by cabbage phospholipase Dcatalyzed transphosphatidylation [24] were purified by HPLC and were kindly donated by Mr. W.S.M. Geurts van Kessel from this laboratory. Phosphatidylserine was isolated from rat brain according to Sanders [25]. Lysophosphatidic acid was synthesized by Crotalus adamanteus phospholipase A, (Sigma, U.S.A.) breakdown of egg phosphatidylcholine and subsequent treatment of the resulting lysophospholipid with phospholipase D from Streptornyces chromofuscus (Boehringer, F.R.G.) as described [26] and was provided by Dr. J.M.M. Kessels from this laboratory. Coenzyme A was obtained from Boehringer, F.R.G. All other chemicals were products of either Merck. F.R.G. or Fluka, Switzerland, and of standard laboratory grade. Methods Isolation of microsomes. Lungs from adult male rats were homogenized with a Potter-Elvehjem homogenizer in 150 mM KCl/50 mM Tris/O.l mM EGTA/ 10 mM P-mercaptoethanol/ 3 mM MgCl Z, pH 7.4 (buffer A) to yield a 10% (w/v) homogenate. After centrifugation at 600 X g for 10 min and at 20000 X g for 20 min, microsomes were spun by centrifugation at 105000 x g for 60 min. The microsomal pellet was resuspended in buffer A to a concentration of 5 mg microsomal protein per ml. Preparation of labeled phosphatidylcholine vesicles. Vesicles were prepared by sonication of 35 PM l-acyl-2-[l-‘4C]linoleoylphosphatidylcholine (22 000 dpm/ nmol) or 1-acyl-2-[ 1-I4 Clarachidonoylphosphatidylcholine (15 000 dpm/ nmol) in the presence of 2 mol% phosphatidic acid in 3 ml

460

of buffer A with a Branson B12 sonifier for 45 min at 50 W under a stream of Nz at 0°C. Subsequently, the sonicate was centrifuged at 105 000 X g for 60 min. Vesicles from the supernatant were used for labeling of the microsomes. Labeling of microsomal membranes with radioactive phosphatidylchofine. 3 ml of the microsomal suspension were mixed with the labeled vesicles in a 1 : 1 ratio (v/v) and incubated for 45 min at 37°C in the presence of 12 pg/ml phosphatidylcholine-exchange protein. After the incubation, microsomes were separated from vesicles by centrifugation at 105 000 X g for 60 min. The microsomal pellet was finally resuspended in buffer A to a concentration of 5 mg microsomal protein per ml. Incubation of microsomes with phospholipids. Prior to incubation, lysophospholipids were sonicated in buffer A for 3 X 1 min at 50 W under N, at 0°C. All incubations of lysophospholipid (amounts specified in the respective figure legends) with microsomes were carried out at 37°C for 30 min in a volume of 1 ml containing 500 pg microsomal protein and 65 PM CoA, unless stated otherwise in the figure legends. Control incubations were performed without CoA. The reaction was stopped by addition of 3 ml CHCl,/CH,OH (1 : 2, v/v). Lipids were extracted according to the method of Bligh and Dyer [27]. After incubations with lysophosphatidic acid, lipids were extracted according to this method after addition of 100 ~1 of 1 M citric acid to the aqueous layer to ensure quantitative recovery of phosphatidate [28]. The lipid extracts were applied onto silica gel type H thin-layer plates. To separate phosphatidylserine and phosphatidylglycerol from phosphatidylcholine, 1 ,l ,l -tricholoroethane/ 2-propanol/ H ,O (45 : 50 : 8, v/v) was used as developing system, whereas for the separation of phosphatidylethanolamine and phosphatidic acid from phosphatidylcholine, the developing systems CHCl,/ CH,OH/CH,COOH/H,O (100 : 50 : 16 : 8, v/) and CHCl,/CH,OH/acetone/CH,COOH/H,O (20 : 4 : 8 : 4 : 2, v/v) were used, respectively. After and exposure to I, vapor, spots were identified scraped into scintillation vials. Analytical procedures. Protein was measured according to the method of Bradford [29] as modified by Vianen and Van den Bosch [30]. Lipid

phosphorus was determined from silica gel as described by Rouser et al. [31]. Radioactivity was measured after addition of Packard emulsifier scintillation fluid in a Packard 3320-Tri-Carb liquid-scintillation spectrometer. Results and Discussion Recently [ll], we demonstrated a CoA-dependent transfer of linoleate from lung microsomal phosphatidylcholine to exogenous I-acyl-2lysophosphatidylethanolamine. In this paper, we describe some characteristics of this reaction in rat lung microsomes. We have applied another technique than previously used by investigators studying this transacylation between phospholipid classes. Most authors labeled the target cells or organelles by incorporation of radiolabeled arachidonate into cellular phospholipids and measured shifts in the distribution of radioactivity upon incubation of membranes in the presence of CoA and/or lysophospholipids [4,6-lo]. We have introduced labeled phosphatidylcholine into the microsomal membrane by incubation with vesicles of radiolabeled phosphatidylcholine in the presence of purified phosphatidylcholine-exchange protein from bovine liver. When an about 40-fold excess of microsomal phosphatidylcholine over vesicle phosphatidylcholine was used, from 45 to 60% of vesicular phosphatidylcholine was transferred to the microsomal membranes to give specific activities of microsomal phosphatidylcholine up to 150 dpm/nmol. Labeling of the microsomal membranes in this way has the advantage that redistribution of radioactivity that might occur during labeling of membrane phospholipids with fatty acids in the presence of CoA can be prevented. Of course, the disadvantage is that only phosphatidylcholine can be studied as acyl donor for transacylase activities. In Fig. 1, the time-dependent transfer of arachidonate from phosphatidylcholine to lysophosphatidylethanolamine is depicted. The non-linear incorporation of arachidonic acid into phosphatidylethanolamine is comparable to the CoA-mediated arachidonoyl transfer from human platelet membrane phospholipids to lysophosphatidylserine, as reported by Kramer et al. [lo]. In rat platelets, Colard et al. [32] showed a transfer-rate of arachidonate from phosphati-

461

i

I

d-l 60

Fig. 1. Time-course phosphatidylcholine amine. Microsomes

of arachidonate to

exogenous

acyl-2.lysophosphatidylethanolamine at 0°C.

Incubations

in a volume of 1 ml, containing PM

CoA

indicated

and 100 PM time

transfer from microsomal Fig. 2. CoA dependency

lysophosphatidylethanol-

somal

were isolated and labeled with I-acyl-2-[l-

“C]arachidonoylphosphatidylcholine. at 50 W under N,

~

intervals,

to incubation,

was sonicated were performed

I-

3 x I min at 37’C

500 pg microsomai protein. 65

lysophosphatidylethanolamine. phosphatidylethanolamine

lated from the total incubation lation vials. Radioactivity

Prior

At the was iso-

mixture and scraped into scintil-

of arachidonate

phosphatidylcholine

ethanolamine.

Radiolabeled

to

microsomes

acyl-2-lysophosphatidylethanolamine

lysophosphatidyl-

and sonicates of l-

were

scribed in the legend of Fig. 1. incubations 37OC for 30 min in a volume

transfer from micro-

exogenous

prepared

as

de-

were carried out at

of 1 ml, containing

500 ftg

microsomal protein, 100 FM lysophosphatidylethanolamineand the

indicated

amine-associated

concentration radioactivity

of

CoA.

Phosphatidylethanol-

was quantitated

as described.

was measured as described.

dylcholine into alkylacylphosphatidylchohne, alkylacylphosphatidylethanolamine and alkenylacylphosphatidylethanolamine that was linear up to 5-6 h. However, these experiments with intact cells were performed without addition of exogenous acceptor lysophosphatides. It is quite conceivable that the transacylation in intact cells remains linear with time for a longer period because of lower amounts of lysophospholipid acceptors. In Fig. 2, the CoA dependency of the transfer of arachidonate from microsomal phosphatidylcholine to exogenous lysophosphatidylethanolamine is shown. The apparent K, value for CoA was 1.5 PM, as calculated from a Hanes plot 1331 of the data. This value for CoA is in good agreement with the apparent K,, value of 1.4 FM for the arachidonoyl transacylation to lysophosphatidylserine in human platelets, as reported by Kramer et al. [lo]. In Fig. 3, the transacylation of arachidonate from phosphatidylcholine to lysophosphatidylethanolamine is plotted against the amount of microsomes in the incubation mixture. In this it is clearly demonstrated that the graph, arachidonoyl transfer increases linearly with the

content of microsomal protein. Since the amount of labeled substrate varies linearly with microsomal proteins in these experiments, the fact that a constant percentage of arachidonate is transferred suggests that the acceptor lysophosphatidylethanolamine is not limiting under the conditions employed. This was borne out in further experiments in which the effect of lysophosphatidylethanolamine concentration was measured. Transacylase activity reached near maximal values above 60 i&M lysophospholipid acceptor (Figs. 4 and 5). In Fig. 4, the enzyme activity with microsomal-associated l-a~yl-2-[l-‘4C]linoleoylphosphatidy~choline and 1-acyl-2-[l -I4 C]arachidonoylphosphatidylcholine was compared. Maximal transfer, as calculated from Hanes plots of the data, amounted to 12.2 t- 1.0% and 2.0 k 0.1% for arachidonate and linoleate, respectively. When transfer activities were calculated from the percent of fatty acyl chains transferred and the specific radioactivities of microsomal phosphatidylcholine at zero time, assuming homogeneity of microsomal phosphatidylcholine, results as depicted in Fig. 4 were obtained. Kinetic parameters as derived from Hanes plots indicated V,,, values for arachidonate and linoleate transfer of 12.0 * 1.0 and 4.6 + 0.2 nmol/30 min per mg protein, respectively. The

0

,

Md~OMAL

540 PROTEIN

,

9&i Lug,

proteins on Fig. 3. Effect of varying amounts of microsomal arachidonate transfer from phosphatidylcholine to lysophosphatidylethanolamine. Radiolabeled microsomes and sonicates of l-acyl-2-lysophosphatidylethanolamine were prepared as described in the legend of Fig. 1. Incubations were carried out at 37°C for 30 min in a volume of 1 ml, containing 100 pM lysophosphatidylethanolamine, 65 PM CoA and the indicated amount of microsomal protein. Phosphatidylethanolamine was isolated and radioactivity measured.

actual amounts, considering the fact that lung phosphatidylcholine contains from 1.5- to 2-fold more linoleate than arachidonate, [34,35] may differ even less. The affinity for lysophosphatidylethanolamine differed also somewhat with the acyl donors, i.e., the apparent K, values for arachidonate and linoleate were 21 k 3 and 38 + 3 PM, respectively. In human platelets, an apparent K,, of 76 PM has been reported by Kramer et al. [lo] for the CoA-mediated arachidonoyl transfer total platelet phospholipids to from lysophosphatidylserine. Using platelets prelabeled with either arachidonate, eicosatrienoate or oleate, these authors found transfer activities of total membrane phospholipids to lysophosphatidylserine corresponding to 7.2, 1.7 and 0.1% of total radioactivity, respectively. Comparing percent transfer of acyl chains in macrophages derived from murine bone marrow, Flesh et al. [8] noticed a linoleoyl transfer from phosphatidylcholine into lysophosphatidylethanolamine that was slightly lower than the transfer of only arachidonate. Oleic acid could be transferred as well, but only with a rate about one-third that of arachidonate. CoA-mediated acyl transfer from endogenous donors to accepting alkyl lysophos-

20 40 60 80 CO120

0 20406D80100120 LPE (,uM)

Fig. 4. Transfer of arachidonate and linoleate from microsomal phosphatidylcholine to exogenous lysophosphatidylethanolamine. Microsomes were labeled with l-acyl-2-[l-‘4C]arachidonoyl- or l-acyl-2-[l-‘4C]linoleoylphosphatidylcholine. Sonicates of lysophosphatidylethanolamine (LPE) were prepared as described in the legend of Fig. 1. Incubations were performed at 37°C for 30 min in a volume of 1 ml, containing 500 pg microsomal protein, 65 PM CoA and lysophosphatidylethanolamine at the indicated concentrations. After extraction, phosphatidylethanolamine was isolated by means of thin-layer chromatography and scraped into scintillation vials. Radioactivity was measured as described. Mean values* S.E. from duplicate experiments are depicted for linoleate transfer (right panel) and arachidonate transfer (left panel). Insets show acyl transfer activity (V) as function of substrate concentration (S) after transformation to Hanes plots.

phatidylcholine in rabbit alveolar macrophage membranes was recently reported to be active with linoleate, arachidonate, oleate and palmitate, in that order [36]. The above results may be indicative of a variable specificity for CoA-mediated acyl transfer reactions in different cells and organelles. In the next experiment, we investigated various lysophosphatides as acceptors for arachidonate from microsomal phosphatidylcholine. As can be seen in Fig. 5, the transacylase system in rat lung microsomes has a striking preference for lysophosphatidylethanolamine at least with phosphatidylcholine as acyl donor. Lysophosphatidylglycerol and lysophosphatidylserine could be to a lower extent than acylated too, albeit lysophosphatidylethanolamine, whereas lysophosphatidate hardly functioned as acceptor. In human platelets [lo], the order of arachidonate acceptance from membrane phospholipids was lysophosphatidylserine > lysophosphatidylethanolamine > lysophosphatidylinositol. These results suggest that CoA-mediated transacylations may exhibit a variable specificity with regard to transferred acyl chains as well as accep-

463

20 40 60 LYSOPHOSPHOLIPID

80

100 (,oM)

Fig. 5. Transfer of arachidonate from microsomal phosphatidylcholine to various acceptor lysophospholipids. Radiolabeled microsomes and sonicates of lysophosphohpid were prepared as described in the legend of Fig. 1. Incubations were carried out at 37°C for 30 min in a volume of 1 ml, containing 500 pg microsomal protein, 65 PM CoA and lysophospholipid at the Indicated concentrations. Phospholipids were extracted, separated and radioactivity quantitated. The transfer rates of from phosphatidylcholine to I-acyl-2arachidonate 0), I-acyl-2lysophosphatidylethanolamine (0 ~ 0). I-acyl-2-lysophosphatilysophosphatidylserine (oA) and I-acyl-2-lysophosphatidylglycerol date (Aare shown from separate experiments. Results are (A -A) given as means f SE. from duplicate experiments.

tor lysophosphatides in different cell types. Our interest in these transacylations originated from the observation [ll] that addition of cytosol to microsomes led to an increase in the percentage of labeled disaturated phosphatidylcholine. The CoA-mediated transfer of arachidonate and linoleate from phosphatidylcholine to lysophosphatidylethanolamine in the cytosol [ll] could contribute to the decrease in the percentage of unsaturated phosphatidylcholine. In lung, as in most other tissues, arachidonate is distributed disproportionately over the different phospholipid classes [35]. i.e., 26.9% of all fatty acids esterified to pulmonary phospholipid is arachidonate acid, whereas phosphatidylcholine contains only about 4% arachidonate [34,35]. Yet, when comparative studies were performed [37-391, acyl-CoA : llysophosphatidylcholine acyltransferase in rat lung microsomes exhibited higher activity with arachidonoyl-CoA than with any of the other acyl-CoAs tested, including those of palmitate, stearate, oleate and linoleate. It is quite feasible that the low arachidonate level in pulmonary phosphatidylcholine at equilibrium is caused by

the specific transfer of arachidonate phosphatidylcholine to other phospholipids by CoA-mediated transacylation, as observed in this paper. It is likely that the lysophosphatidylcholine produced is subsequently reacylated either by acylCoA : lysophosphatidylcholine acyltransferases or by CoA-mediated transacylation with an as yet unknown acyl donor. In this respect, it is interesting to note that a transacylation of palmitate, esterified at the sn-2-position of phosphatidylglycerol, to 1-acyl-2-lysophosphatidylcholine has been observed in dog lung microsomes and type II pneumocyte-derived cultures [40]. The mechanism of this acyl transfer could not be assessed definitively at that time and because of the low activity of the putative phosphatidylglycerol : lysophosphatidylcholine transacylase in comparison to acyl-CoA : lysophosphatidylcholine acyltransferase, the former was concluded not to be a major factor in the biosynthesis of pulmonary dipalmitoylphosphatidylcholine [40]. It has to be noted, however, that the transacylase activity in dog lung microsomes was measured in the absence of added CoA. Earlier experiments have indicated that palmitate derived de novo in lung from acetate gave a higher specific activity in phosphatidylglycerol than in phosphatidylcholine [41]. These results leave the possibility for a precursor-product relationship to exist between palmitate in phosphatidylglycerol and that in phosphatidylcholine, possibly through CoA-mediated transacylation. The actual occurrence of this reaction and its contribution to pulmonary phosphatidylcholine-remodeling remains to be established. Acknowledgements

This study was carried out under the auspices of the Netherlands Foundation for Chemical Research (S.O.N.) with financial aid from the Netherlands Organization for the Advancement of Pure Research (Z.W.O.). References 1 Irvine, R.F. (1982) Biochem. J. 204, 3-16 2 Hill, E.E. and Lands, W.E.M. (1970) in Lipid Metabolism (Wakil, S.J., ed.), pp. 185-279, Academic Press, New York 3 Holub. B.J. and Kuksis. A. (1978) Adv. Lipid Res. 16, l-125

4 Irvine, R.F. and Dawson, R.C.M. (1979) Biochem. Biophys. Res. Commun. 91, 139991405 5 Van den Bosch, H.. Van Golde. L.M.G. and Van Deenen, L.L.M. (1972) Rev. Physiol. 66. 13-145 6 Trotter. J. and Ferber. E. (1981) FEBS Lett. 128. 237-241 7 Trotter. J.. Flesh, I., Schmidt, B. and Ferber, E. (1982) J. Biol. Chem. 257, 1816-1823 8 Flesh. I., Ecker, B. and Ferber, E. (1984) Eur. J. Biochem. 139, 431-437 9 Kramer, R.M. and Deykin. D. (1983) J. Biol. Chem. 258. 13806-13811 10 Kramer, R.M., Pritzker, C.R. and Deykin, D. (1984) J. Biol. Chem. 259, 2403-2406 11 Nijssen, J.G. and Van den Bosch. H. (1985) 875, 450-457 12 Ohta. M., Hasegawa, H. and Ohno, K. (1972) Biochim. Biophys. Acta 280, 552-558 13 Garcia, A., Newkirk. J.D. and Mavis. R.D. (1975) Biochem. Biophys. Res. Commun. 64, 128-135 14 Longmore. W.J., Oldenborg, V. and Van Golde, L.M.G. (1979) Biochim. Biophys. Acta 572, 452-460 15 Vereyken. J.M.. Montfoort. A. and Van Golde. L.M.G. (1972) B&him. Biophys. Acta 260, 70-81 16 Lands, W.E.M. (1960) J. Biol. Chem. 235, 2233-2237 17 Frsolono. M.F., Slivka. S. and Charms. B.J. (1971) J. Lipid Res. 12.96-103 18 Tansey, F.A. and Frosolono, M.F. (1975) Biochem. Biophys. Res. Commun. 67, 1560-1566 19 Snyder. F. and Malone, B. (1975) Biochem. Biophys. Res. Commun. 66, 914-919 20 Batenburg. J.J., Longmore. W.J.. Klazinga, W. and Van Golde, L.M.G. (1979) Biochim. Biophys. Acta 573, 136-144 21 Van den Bosch. H. (1980) Biochim. Biophys. Acta 604, 191-246 22 Van den Bosch, H.. Aarsman. A.J. and Van Deenen, L.L.M. (1974) Biochim. Biophys. Acta 348. 197-209

23 Westerman, J., Kamp, H.H. and Wirtz, K.W.A. (1983) Methods Enzymol. 98. 581-586 24 Comfurius, P. and Zwaal. R.F.A. (1977) Biochim. Biophys. Acta 488, 36-42 25 Sanders, H. (1976) Biochim. Biophys. Acta 144, 485-487 26 Imamura, S. and Horiuti, Y. (1979) J. Biochem. 85. 79-95 27 Bligh. E.G., and Dyer, W.J. (1959) Can. J. Biochem. 37. 911-917 28 Bjerve. K.S., Daae, L.N.W. and Bremer, J. (1974) Anal. Biochem. 58, 238-245 29 Bradford, M.M. (1976) Anal. Biochem. 72. 248-254 30 Vianen, G.M. and Van den Bosch, M. (1978) Arch. Biothem. Biophys. 190, 373-384 31 Rouser, G., Siakotos. A.N. and Fleisher. S. (1966) Lipids 1. 85-86 32 Colard, 0.. Breton. M. and Bereziat, G. (1984) B&hem. J. 222. 657-662 33 Hanes. C.S. (1932) Biochem. J. 26, 140661421 34 Ohno, K.. Akino, T. and Fujiwara. T. (1978) in Reviews in Perinatal Medicine (Scarpelli, E.M. and Cosmi, E.V.. eds.), Vol. 2. pp. 2277318, Raven Press. New York 35 Alam, S.Q. and Alam, B.S. (1984) Lipids 19, 38-43 36 Robinson. M.. Blank, M.L. and Snyder, F.,(1985) J. Biol. Chem. 260. 7889-7895 37 Hasegawa-Sasaki, H. and Ohno. K. (1975) Biochim. Biophys. Acta 380, 486-495 38 Yamada, K. and Okyyama. H. (1979) Arch. Biochem. Biophys. 196, 209-219 39 Holub, B.J.. Piekarski, J. and Possmayer. F. (1980) Can. J. Biochem. 58, 434-439 40 Stanford, G.L. and Frosolono. M.F. (1981) Biochim. Biophys. Acta 665, 339-344 41 Sauders, R.L. and Longmore, W.J. (1975) Biochemistry 14. 8355840